Dear all,
I am attempting to estimate the effect of the change in temperature and precipitation (at the district level) and their interactions with historical means (HT, Hp respectively) on the changes in households’ consumption. My FD model looks as follow:

ΔCh = b0 + b1ΔTd + b2* ΔTd *HTd + b3 ΔPd + b4 ΔPd*HPd +eh

b0 captures the time trends. I cluster the standard errors at the district level (i.e. level of treatment).

I came across the paper de Chaisemartin and D’Haultfoeuille (2018): Two-way fixed effects estimators with heterogeneous treatment effects.
I tried to do their suggested diagnosis using the command twowayfeweights to check whether, my FD estimation suffers from negative weights.
On the first sight, it seems that the negative weights are indeed a problem, though I am still not sure whether I applied the syntax correctly.
I am puzzled, because I do not only have one treatment, but several independent variables whose effect I am interested in.
What would be the correct way to run the diagnosis command: twowayfeweights?

Moreover, I am also not sure how to run the fuzzydid correctly to obtain the Wald TC estimator. Could you help me here?
What would be in this case my control group?


Many thanks.